Arrays of small speakers

Archived from groups: rec.audio.tech (More info?)

I remember seeing information on arrays of inexpensive small speakers
(multiple small drive units in a single enclosure) for high-fidelity
purposes. I think this was published in magazines in the 1950s or 60s; I
have the impression that Bose started out producing speaker systems this
way.

I don't remember and can't find the references. Can anybody please point
me to information on this topic, either from that period or later?

adthanksvance,
--
Michael Salem

Archived from groups: rec.audio.tech (More info?)

Stu-R wrote:

> The speaker you're looking for was called the "Sweet Sixteen". It was
> published in popular electronics. Check the following link.
>
> http://www.eagle-wing.net/ClickPic [...] nics.shtml

Many thanks!

'"Sweet Sixteen" loudspeaker system, published in the January 1961 issue
of Popular Electronics.'

I don't think this is the actual article I remember, which I believe had
a lot of really small speakers, but of course I'm looking for anything
that's there, not just what I read.

Best wishes,
--
Michael Salem

Archived from groups: rec.audio.tech (More info?)

Michael Salem wrote:

> I remember seeing information on arrays of inexpensive small speakers
> (multiple small drive units in a single enclosure) for high-fidelity
> purposes. I think this was published in magazines in the 1950s or
> 60s; I have the impression that Bose started out producing speaker
> systems this way.
>
> I don't remember and can't find the references. Can anybody please
> point me to information on this topic, either from that period or
> later?

Sure, there was a Popular Electronics article about a project called "The
Sweet 16". It was composed of 16 el-cheapo AM radio speakers in a minimal
box. This was around 1960. I heard one at the time in a local electronics
store, and thought it was some kind of a sonic disaster. It realized every
cliché about trying to make a silk purse out of a pig's ear etc. that you
ever heard. Later on there was a sequel project called "The Sweet 16+1". It
added a cheap tweeter. This was basically trying to put lipstick on a pig,
polish a turd, you name it.

In the day of "The Sweet 16", remarkably little was generally known about
loudspeakers and loudspeaker systems, compared to what is known today. So,
it can be excused on the grounds of general ignorance. The first problem was
that the speakers the Sweet 16" was made out, of were systematically
low-fi. The article argued that by combining many of them, the random
variations would be evened out. In fact the speaker drivers were
consistently low-fi.

Then there was the problem of the Sweet 16" speaker array itself. When
speakers are placed close together, a variety of complex interactions
result. These tend to add many more frequency response variations on top of
the many that might be present in the individual drivers themselves.

It turns out that there is a similar array that can work, but it has 25
speakers, not 16. It's called a Bessel array N=25, and details relating to
is can be found in some posts I made in the past few months. I recently
built a Bessel Array N=5, and it works, but.

When all is said and done, the Bessel array N=25 does not deliver 25 times
the sound of one of the speakers that it is composed of. Some of the drivers
must be connected with reversed polarity. The performance of about a
quarter of the array is sacrificed to make the rest of the array work
reasonably well. In the final analysis, you end up with a system that
performs pretty much like just one of the drivers if you sit some distance
from it. Close up is still not a pretty picture. However it does get
considerably louder if you apply much more power.

The economics of driver arrays are not always the best. It's generally
easier and more practical to get more loudness with a larger
high-performance driver, than with many little ones. Arrays of drivers can
be justified when the largest drivers generally available are used to build
them. The costs of larger higher performance drivers have fallen
considerably, and therefore they should be used whenever possible.

There are some very interesting things that can be done with driver arrays
when they are driven by complex systems of power amplifiers and signal
processors. But, they are justified only when simpler systems composed of
fewer drivers, fewer amplifiers and simpler signal processors can't do the
job.

Archived from groups: rec.audio.tech (More info?)

On 5 Jun 2004 04:35:55 -0700, dlsalter@paonline.com (Lee Salter)
wrote:

>I think that probably the bigest difference is that the Sweet 16 was
>done experimentally and Bose actually applied physics and advanced
>math in designing their products.

I've always wondered how Bose could get things so nearly right with
their PA speakers - the 802 etc. - and so horribly wrong with their
domestic stuff.

Archived from groups: rec.audio.tech (More info?)

Laurence Payne wrote:
> On 5 Jun 2004 04:35:55 -0700, dlsalter@paonline.com (Lee Salter)
> wrote:
>
>> I think that probably the bigest difference is that the Sweet 16 was
>> done experimentally and Bose actually applied physics and advanced
>> math in designing their products.
>
> I've always wondered how Bose could get things so nearly right with
> their PA speakers - the 802 etc. - and so horribly wrong with their
> domestic stuff.

What exactly do you find "right" about hte 802 ?


geoff

Archived from groups: rec.audio.tech (More info?)

On Sun, 6 Jun 2004 12:50:50 +1200, "Geoff Wood"
<geoff@paf.co.nz-nospam> wrote:

>> I've always wondered how Bose could get things so nearly right with
>> their PA speakers - the 802 etc. - and so horribly wrong with their
>> domestic stuff.
>
>What exactly do you find "right" about hte 802

They were standard issue for theatre sound reinforcement in UK
theatres a generation ago. Did a pretty good job, as I remember.

Archived from groups: rec.audio.tech (More info?)

On Tue, 08 Jun 2004 21:16:40 +0200, Per Stromgren
<per.stromgren@telia.com> wrote:

>On Fri, 4 Jun 2004 17:38:59 +0100, Michael Salem <a$-b$1@ms3.org.uk>
>wrote:
>>I remember seeing information on arrays of inexpensive small speakers
>>(multiple small drive units in a single enclosure) for high-fidelity
>>purposes.
>
>The problem with this approach is the displacement. Altough it seems
>like a good idea to use many small drivers instaed of one big, the
>small still have low xmax compared to the big cone, menaing that the
>max SPL is a lot lower than a single driver with the same area.

The second major problem is that a linear array of speakers is a comb
filter.

Pat http://www.pfarrell.com/prc/

Archived from groups: rec.audio.tech (More info?)

As OP, many thanks for all the responses to my question. The conclusion
I reach from the postings and URLs is that a lot of terrible speaker
drive units combine, if carefully built, to form one cheap, but
mediocre, speaker.

Another project of the same sort that I actually built was to use a
couple of small (2"?), cheap, transistor-radio speakers in some sort of
enclosures (I used cylindrical plastic food containers) to make cheap
headphones which were supposed to punch above their weight (I.e., sound
as good as much more expensive headphones). I remember being satisfied,
but that might just have been influenced by pride in my own handiwork...

Best wishes,
--
Michael Salem

Archived from groups: rec.audio.tech (More info?)

Per Stromgren wrote:
> On Fri, 4 Jun 2004 17:38:59 +0100, Michael Salem <a$-b$1@ms3.org.uk>
> wrote:
>
>> I remember seeing information on arrays of inexpensive small speakers
>> (multiple small drive units in a single enclosure) for high-fidelity
>> purposes.
>
> The problem with this approach is the displacement.

Another problem is that is many times less a point-source thn even a 3-way
speaker.

geoff